Wednesday, October 7, 2020

The Electoral College Map (10/7/20)

Update for October 7.

On the day of the one and only vice presidential debate of the 2020 cycle, a veritable flood of new state-level polling data was released, further refining the overall picture of the race. Note that that says refine and fundamentally reshape. While there were some attention-grabbing numbers in a few of these polls -- especially the trio from Quinnipiac -- most of the new data only served to maintain the status quo here at FHQ. And while there were subtle changes, they tended to be margins that shifted in Biden's favor rather than states or districts switching categories and/or jumping the partisan line altering the projected electoral vote tally. 

And honestly, that is to be expected with the graduated weighted average that FHQ has utilized for four presidential election cycles now. If a race is steady like 2012 was (and to some lesser degree 2020 is), then the average will guard against any wild fluctuations created by temporary polling changes. It was around this time in 2012 that Romney slipped into the lead in national polls. But at FHQ the projection never changed. Florida got close to switching into Romney territory after that first debate, but never quite did and stayed in Obama's column down the stretch. 

Now, whether 2020 remains steady like 2012 remains an open question. What FHQ calls temporary polling changes above may not prove to be all that temporary. It could be evidence of the dam beginning to break on the president. But with 27 days until election day, there should be enough time and enough polling where it matters to bring the graduated weighted averages along. In the best case, things break late here, confirming any long term changes between now and election day. The worst case looks more like the misses in 2016, when the shift occurred late as a larger pool of undecideds broke and state-level polling never really reflected that. 

In any event, the mantra around here with respect to the averages is that when a change occurs in the polling, it tends to create a lasting change in the averages. 

On to the polls...

Polling Quick Hits:
(Biden 48, Trump 43 via Data Orbital | Biden 48, Trump 46 via Ipsos)
[Current FHQ margin: Biden +3.21]
Both Data Orbital and Ipsos conducted polls in Arizona in mid-September, but only the survey from Data Orbital saw any significant change in the time since then. While Biden remained steady, Trump share of support dropped off just below his average share of support at FHQ. But both of these surveys were on target on Biden's current average level of support in the state.

(Biden 49, Trump 44 via Cherry Communications | Biden 51, Trump 40 via Quinnipiac | Biden 49, Trump 45 via Ipsos)
[Current FHQ margin: Biden +3.48]
The Sunshine state saw a trio of releases on the day and all were from firms who had previously been in the field there. The attention is clearly going to gravitate toward the Quinnipiac survey where Biden grew his support while Trump saw his decline. But this latest Q-poll look a lot more like the survey the school released in July during Biden's polling surge. Does this hail a return to that? It may be a signal, but it could also be an outlier. That will not become clear until more polling is done. The trajectory of the movement in the other two surveys also moved in Biden's direction but in a much more muted way than through the Quinnipiac lens. Florida's average margin fell into the threes, but has plateaued there recently and may even be reversing course some. 

(Biden 48, Trump 47 via Civiqs | Biden 50, Trump 45 via Quinnipiac)
[Current FHQ margin: Trump +1.10]
Further north in the Hawkeye state, the margin may still narrowly favor the president, butBiden led in a couple of new surveys. This is the first time that Quinnipiac has been in the field in Iowa, but it has the former vice president ahead by his largest margin there all cycle (minus a June Binder poll with only 200 respondents where Biden was up six). Like the Florida Q-poll above, this one should be taken with a grain of salt for the time being. It may be a harbinger of things to come, but could also just be overly rosy for the Democratic nominee. Civiqs had conducted an Iowa poll before and during June in the window of Biden's surge. The poll then was was a dead heat, and this latest one is as well, although Biden has the narrow edge. This one is more in line with where FHQ's averages have the race currently pegged,

(Biden 61, Trump 32)
[Current FHQ margin: Biden +31.72]
One could split hairs and say that the Change Research survey of Maryland was off the mark in understating Biden's share of support in the Old Line state, but it was only by a couple of points. Otherwise, this one looks a lot like where the race currently stands for Maryland's ten electoral votes. 

(Biden 51, Trump 43)
[Current FHQ margin: Biden +7.10]
Ipsos was also in the field in Michigan for the first time since mid-September and found a race that had moved in Biden's direction but only modestly. The former vice president gained a couple of points to crack 50 percent again. That mark is important because it would close the door on the state's 16 electoral votes if Biden can get there on election day. Of the 16 polls conducted since that last Ipsos poll of the Great Lakes state, 12 have had the former vice president over the majority mark. Furthermore, he is approaching that level in the averages. 

(Trump 50, Biden 48)
[Current FHQ margin: Trump +6.37]
The only other time Garin-Hart-Yang were in the field in the Show-Me state was back in June. And although that falls in the period during Biden's rise in the polls across the country, it was also the only time in calendar 2020 that a survey had the Democratic nominee in the lead in Missouri. What was an outlier then is probably a little less so now. But this one still has Trump running just below his average there and Biden more than three points above his. Missouri has already shown to be be closer in 2020 than in 2016, but likely not this close. 

(Biden 48, Trump 42)
[Current FHQ margin: Biden +3.62]
Outside of the two Fox News polls, this Siena/NYT Upshot survey has Biden up by his largest margin in the Silver state in calendar 2020. But this one is also not all that inconsistent with the firm's September poll in underpolled Nevada. Biden's share of support rose by a couple of points to match his FHQ average share of support. The president, meanwhile, remained at 42 percent, lagging a couple of points behind his. The introduction of the four waves of Survey Monkey polls over the weekend took Nevada off the Watch List, but polls like these will have it tracking back in that direction. 

North Carolina
(Biden 47, Trump 47)
[Current FHQ margin: Biden +1.51]
FHQ has gotten in the habit of saying North Carolina is close. And it is. This Ipsos poll does little to dislodge the Tar Heel state from that category. Nor does the fact that the race has been stable in the time since the firm's September poll of the state. A race knotted at 47 then is the same now. Neither diverges much from the 47-46 (rounded) advantage Biden consistently holds there. 

(Biden 45, Trump 44)
[Current FHQ margin: Trump +0.40]
This is the first time Siena/NYT Upshot has ventured into the Buckeye state to gauge preferences in the race for the state's 18 electoral votes. What the firm found was a close race with Biden marginally ahead. That differs from the current margin that continues to show the president in the lead by a sliver of a point, and does so with both candidates running below their established FHQ average shares of support. But that is attributable to the still high number of undecideds, a common thread in these Siena polls conducted in conjunction with the Upshot. Despite that, this survey nudged Ohio a bit closer to the partisan line. 

(Biden 51, Trump 47 via Emerson | Biden 54, Trump 41 via Quinnipiac)
[Current FHQ margin: Biden +5.49]
Like several other states, Pennsylvania, too, had a couple of poll releases on the day. Both were repeat offenders, having previously conducted surveys in the commonwealth. Unlike the other two states with Quinnipiac polls, the change poll over poll in the Keystone state was not to the same degree. In fact, there has been one common thread across the three Q-polls of Pennsylvania in 2020: they have all had Biden at or above 50 percent and all have had Trump in the low 40s. The lead certainly widened since the early September poll, but not like in Florida (even if both new polls resembled one another). The Emerson update looks similar to the August survey, but this one included third party candidates. That aside, Trump stayed at 47 percent, but Biden dropped a couple of points.

(Biden 48, Trump 48 via Civiqs | Biden 49, Trump 49 via EMC Research)
[Current FHQ margin: Trump +1.61]
Both new surveys in the Lone Star state had the two candidates running above their established average shares of support but also showed a decreasing number of undecideds. That continues to be something to watch as election day nears: where those undecided respondents end up. Biden has had polling leads in Texas throughout the summer, but if undecideds smooth things out for both candidates and draw this race closer to a tie, then that is a potentially huge development with 38 electoral votes on the line. That the Trump campaign and Republicans are being made to spend in Texas at all (or simply rely on past Republican support to get Trump over the line without the spending) says much about the state of this race for the White House. Again, if Texas is among the most competitive states on election day, then Biden is likely sitting pretty in his quest for 270.

West Virginia
(Trump 56, Biden 38)
[Current FHQ margin: Trump +30.11]
Much has and will be made of the 18 point Trump advantage in the new Triton Polling and Research of the Mountain state. And on the surface, it is jarring. The president won West Virginia by 42 points in 2016, so a 24 point swing is no small thing. But it is a massive outlier if things look like this on election day. The current average swing from the 2016 election to polling now is about six and three-quarters points toward the Democrats. This sort of change in West Virginia would be way above average. But there have been clunker polls elsewhere in red states in 2020 as well. The four point Trump lead in Alabama from Tyson Group in August comes to mind. One should expect the margin in West Virginia to be closer in 2020 than in 2016 on the notion of a uniform swing toward the Democrats alone. But it likely will not be quite this acute.

(Biden 47, Trump 42)
[Current FHQ margin: Biden +6.12]
Last but not least, solid Marquette Law School updated the state of the race in the Badger state for the first time since early September and continued to show a steady race. Both candidates lost a little off of their September support, but Biden slightly increased his advantage in that time. This one is not far off the average margin, but it does have both candidates falling short of their average shares of support, but Biden a bit more so. Trump will not need Michigan, Pennsylvania and Wisconsin to get to 270, but he will need one of them and Pennsylvania sits closer to the partisan line right now than the other two. But it remains more than five points out of the president's grasp and is drifting further away. 

NOTE: A description of the methodology behind the graduated weighted average of 2020 state-level polling that FHQ uses for these projections can be found here.

The Electoral College Spectrum1
(273 | 285)
(279 | 265)
(308 | 259)
(319 | 230)
NE CD1-1
ME CD2-1
(335 | 219)
ME CD1-1
NE CD2-1
NE CD3-1
1 Follow the link for a detailed explanation on how to read the Electoral College Spectrum.

2 The numbers in the parentheses refer to the number of electoral votes a candidate would have if he or she won all the states ranked prior to that state. If, for example, Trump won all the states up to and including Pennsylvania (Biden's toss up states plus the Pennsylvania), he would have 285 electoral votes. Trump's numbers are only totaled through the states he would need in order to get to 270. In those cases, Biden's number is on the left and Trump's is on the right in bold italics.

3 Pennsylvania
 is the state where Biden crosses the 270 electoral vote threshold to win the presidential election, the tipping point state. The tipping point cell is shaded in yellow to denote that and the font color is adjusted to attempt to reflect the category in which the state is.

There was so much to look at polling-wise on the day that debate season resumes. However, there was nary a change across the graphics at FHQ. The map and Watch List -- those states within a fraction of a point of changing categories -- remained exactly as they were yesterday. Only Missouri, on the strength of that new GHY survey, pushed past Alaska and closer to the partisan line separating both candidates coalitions of states. But that group of Alaska, Missouri and South Carolina are fairly tightly clustered on the lower end of the Lean Trump category, but just off the Watch List. None of the three are likely to fall to the Democrats unless the bottom truly drops out on Trump, but that Lean/Toss Up line on Trump's side is probably the cut off point. Texas is potentially achievable for Biden but it is the last state in the order that is likely to flip blue in November. That is not to say that Texas will or will not flip, but it is likely Biden's high water mark in a landslide scenario (again, unless the dam breaks on Trump).

27 days to go.

Where things stood at FHQ on October 7 (or close to it) in...

NOTE: Distinctions are made between states based on how much they favor one candidate or another. States with a margin greater than 10 percent between Biden and Trump are "Strong" states. Those with a margin of 5 to 10 percent "Lean" toward one of the two (presumptive) nominees. Finally, states with a spread in the graduated weighted averages of both the candidates' shares of polling support less than 5 percent are "Toss Up" states. The darker a state is shaded in any of the figures here, the more strongly it is aligned with one of the candidates. Not all states along or near the boundaries between categories are close to pushing over into a neighboring group. Those most likely to switch -- those within a percentage point of the various lines of demarcation -- are included on the Watch List below.

The Watch List1
Potential Switch
from Toss Up Trump
to Toss Up Biden
New Hampshire
from Strong Biden
to Lean Biden
New Mexico
from Strong Biden
to Lean Biden
from Toss Up Trump
to Toss Up Biden
from Lean Biden
to Toss Up Biden
1 Graduated weighted average margin within a fraction of a point of changing categories.

Methodological Note: In past years, FHQ has tried some different ways of dealing with states with no polls or just one poll in the early rounds of these projections. It does help that the least polled states are often the least competitive. The only shortcoming is that those states may be a little off in the order in the Spectrum. In earlier cycles, a simple average of the state's three previous cycles has been used. But in 2016, FHQ strayed from that and constructed an average swing from 2012 to 2016 that was applied to states. That method, however, did little to prevent anomalies like the Kansas poll that had Clinton ahead from biasing the averages. In 2016, the early average swing in the aggregate was  too small to make much difference anyway. For 2020, FHQ has utilized an average swing among states that were around a little polled state in the rank ordering on election day in 2016. If there is just one poll in Delaware in 2020, for example, then maybe it is reasonable to account for what the comparatively greater amount of polling tells us about the changes in Connecticut, New Jersey and New Mexico. Or perhaps the polling in Iowa, Mississippi and South Carolina so far tells us a bit about what may be happening in Alaska where no public polling has been released. That will hopefully work a bit better than the overall average that may end up a bit more muted.

Related posts:

No comments: